INDEX

Explanations

words indicating additional information or referencing further reading material

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "further" along with references to additional information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 hers

-0.74

Ïī

-0.67

 theirs

-0.66

 snatch

-0.65

 paddle

-0.64

iren

-0.64

endered

-0.62

 slack

-0.62

 fishing

-0.61

itably

-0.61

POSITIVE LOGITS

 Examples

1.03

 Locations

1.00

 Reports

0.97

 Stories

0.93

 Quote

0.93

 Advice

0.93

 Problems

0.93

 Issues

0.92

 Thoughts

0.92

 Analysis

0.92

Activations Density 0.071%

words indicating additional information or referencing further reading material

occurrences of the term "further" along with references to additional information

No Comments

No Known Activations