INDEX

Explanations

comparisons indicating a lack of quality or quantity

oai_token-act-pair · gpt-3.5-turbo

phrases expressing relative inadequacy or failing to meet expectations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itton

-0.66

 unsure

-0.62

not

-0.58

olla

-0.57

FL

-0.55

 Mayhem

-0.54

Ĥİ

-0.54

 shortly

-0.53

maybe

-0.53

unknown

-0.53

POSITIVE LOGITS

 anymore

1.15

 sufficiently

0.98

icably

0.98

 enough

0.93

 terribly

0.93

 adequately

0.87

 Enough

0.85

 sufficient

0.85

 necessarily

0.84

enough

0.83

Activations Density 0.332%

comparisons indicating a lack of quality or quantity

phrases expressing relative inadequacy or failing to meet expectations

No Comments

No Known Activations

comparisons indicating a lack of quality or quantity

phrases expressing relative inadequacy or failing to meet expectations

No Comments

No Known Activations