INDEX

Explanations

onyms or variations of the word "go"

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "go."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Aram

-0.70

 certific

-0.61

 ammon

-0.60

picking

-0.60

 uncommon

-0.59

lake

-0.58

rican

-0.58

 accur

-0.56

 throats

-0.56

 elbows

-0.56

POSITIVE LOGITS

vernment

1.11

verning

1.00

ogly

0.92

ctors

0.88

ven

0.87

lems

0.86

ardless

0.84

ption

0.83

eker

0.82

etz

0.82

Activations Density 0.012%

onyms or variations of the word "go"

the repeated use of the word "go."

No Comments

No Known Activations