INDEX

Explanations

terms related to regularly occurring actions or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "regularly."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

lda

-0.86

IDA

-0.85

tein

-0.72

elo

-0.71

 Millennium

-0.70

hemy

-0.69

 Dign

-0.69

eer

-0.65

 Disease

-0.65

arta

-0.64

POSITIVE LOGITS

 spaced

1.04

 occas

0.87

 encountered

0.85

 encount

0.84

 reapp

0.83

 replen

0.81

 scheduled

0.80

 fluct

0.79

 stocked

0.78

 destro

0.78

Activations Density 0.008%

terms related to regularly occurring actions or events

occurrences of the word "regularly."

No Comments

No Known Activations