INDEX

Explanations

phrases related to harmful incidents or accidents involving individuals

oai_token-act-pair · gpt-3.5-turbo

references to male individuals and their actions or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

endif

-0.89

peak

-0.78

Interest

-0.69

 Dominion

-0.68

soType

-0.62

 Crusade

-0.61

icion

-0.59

 Revolution

-0.59

 Rapt

-0.59

problem

-0.58

POSITIVE LOGITS

'd

1.11

 underwent

1.03

 unsuccessfully

1.00

 participated

1.00

 feared

0.99

 witnessed

0.97

 received

0.96

 testified

0.93

 alleges

0.92

 encountered

0.91

Activations Density 0.229%

phrases related to harmful incidents or accidents involving individuals

references to male individuals and their actions or experiences

No Comments

No Known Activations