INDEX

Explanations

proper nouns related to current events and politics

oai_token-act-pair · gpt-3.5-turbo

references to significant political entities and events

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

tnc

-0.80

EStreamFrame

-0.68

 guiActiveUnfocused

-0.67

î

-0.64

 behalf

-0.63

 externalToEVAOnly

-0.61

nants

-0.59

COMPLE

-0.59

-0.58

ãģĨ

-0.57

POSITIVE LOGITS

 meanwhile

0.80

 thri

0.69

isn

0.68

 crept

0.67

 Slate

0.66

 reacted

0.65

 analogy

0.65

 sucks

0.65

 hasn

0.65

 grew

0.65

Activations Density 0.864%

proper nouns related to current events and politics

references to significant political entities and events

No Comments

No Known Activations