INDEX

Explanations

the word "as" followed by a phrase indicating cause or effect

oai_token-act-pair · gpt-3.5-turbo

the word "as" 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fml

-0.72

eur

-0.67

)))

-0.66

Build

-0.65

itiveness

-0.62

tick

-0.62

ishly

-0.61

BACK

-0.61

âĸº

-0.61

åħ

-0.60

POSITIVE LOGITS

 evidenced

1.64

 opposed

1.36

 well

1.15

 illustrated

1.14

 demonstrated

1.10

 shown

1.06

 recounted

0.98

 indicated

0.96

 witnessed

0.96

 exempl

0.95

Activations Density 0.069%

the word "as" followed by a phrase indicating cause or effect

the word "as" in various contexts

No Comments

No Known Activations