INDEX

Explanations

mentions of support for a specific organization or cause

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "that" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Doctrine

-0.67

 Corps

-0.62

Ruk

-0.62

 Hallow

-0.61

Sad

-0.61

NES

-0.60

 Anyway

-0.59

raz

-0.59

 Planning

-0.59

riot

-0.58

POSITIVE LOGITS

 arose

1.01

 preceded

0.93

 comprise

0.91

 occur

0.89

 accumulate

0.89

 weren

0.88

 arise

0.88

 resulted

0.88

 circulate

0.86

 compose

0.86

Activations Density 0.191%

mentions of support for a specific organization or cause

the presence of the word "that" in various contexts

No Comments

No Known Activations