INDEX

Explanations

phrases describing categories or types of things

oai_token-act-pair · gpt-3.5-turbo

phrases that describe different types or categories of things

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

-0.79

rity

-0.69

autions

-0.68

 furthermore

-0.67

 otherwise

-0.67

 evidently

-0.66

 Matters

-0.66

ULTS

-0.66

ourses

-0.66

 promptly

-0.64

POSITIVE LOGITS

 Frankenstein

0.86

 inverse

0.82

 shorthand

0.82

 Trojan

0.81

 sponge

0.79

 miniature

0.78

 cottage

0.78

 Craigslist

0.77

 glue

0.76

 precursor

0.75

Activations Density 0.322%

phrases describing categories or types of things

phrases that describe different types or categories of things

No Comments

No Known Activations