INDEX

Explanations

proper nouns or names of individuals

oai_token-act-pair · gpt-3.5-turbo

mentions of significant events, arrests, and consequences in societal 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

!.

-0.69

inis

-0.65

}.

-0.64

+.

-0.60

};

-0.59

cellaneous

-0.57

utterstock

-0.57

''.

-0.56

.$

-0.56

.''

-0.55

POSITIVE LOGITS

 lacks

0.70

 hadn

0.70

 lacked

0.69

 should

0.69

 shouldn

0.66

 cannot

0.65

had

0.64

 hasn

0.60

 behaved

0.60

 exists

0.58

Activations Density 0.993%

proper nouns or names of individuals

mentions of significant events, arrests, and consequences in societal contexts

No Comments

No Known Activations