INDEX

Explanations

mentions of physical locations or settings

oai_token-act-pair · gpt-3.5-turbo

occurrences of the article "a" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Edit

-0.63

Contents

-0.61

Att

-0.58

 Jagu

-0.57

Events

-0.57

Catal

-0.55

agree

-0.54

evidence

-0.54

tests

-0.53

Hon

-0.53

POSITIVE LOGITS

lot

0.99

 bunch

0.96

 handful

0.89

 plethora

0.88

rouse

0.88

few

0.85

 couple

0.84

 huge

0.82

 sizeable

0.82

 multitude

0.82

Activations Density 0.852%

mentions of physical locations or settings

occurrences of the article "a" in various contexts

No Comments

No Known Activations