INDEX

Explanations

phrases or segments related to news articles or reports

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "which" to identify descriptions or explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Behind

-0.73

cup

-0.73

athi

-0.72

uta

-0.69

rollers

-0.67

 Solution

-0.67

ben

-0.65

die

-0.64

angu

-0.63

nor

-0.60

POSITIVE LOGITS

 culmin

0.95

 comprises

0.93

 lasted

0.92

 consisted

0.88

 prompted

0.88

 culminated

0.88

 consists

0.87

 amounted

0.85

 resulted

0.84

 originated

0.82

Activations Density 0.104%

phrases or segments related to news articles or reports

phrases that include the word "which" to identify descriptions or explanations

No Comments

No Known Activations