INDEX

Explanations

quotations and statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

quotes or speech from characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

rats

-0.75

ģ«

-0.67

 effected

-0.67

respective

-0.61

 tainted

-0.60

impact

-0.60

levant

-0.60

ä»£

-0.57

 hijacked

-0.57

corruption

-0.56

POSITIVE LOGITS

 chuck

0.90

 recalls

0.84

 laughs

0.80

Laughs

0.79

 remembers

0.78

 pauses

0.75

 recalling

0.74

laughs

0.73

 shrug

0.69

toggle

0.69

Activations Density 0.653%

quotations and statements made by individuals

quotes or speech from characters

No Comments

No Known Activations