INDEX

Explanations

information being presented or shared with an emphasis

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "presented."

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

leans

-0.67

illard

-0.66

sei

-0.63

kowski

-0.62

bia

-0.62

assin

-0.61

ne

-0.61

lex

-0.61

profit

-0.61

si

-0.60

POSITIVE LOGITS

eering

0.98

 presentations

0.94

ocument

0.93

 presented

0.91

 Present

0.89

 presents

0.86

escription

0.84

ĸļ

0.83

Present

0.82

emonium

0.81

Activations Density 0.021%

information being presented or shared with an emphasis

occurrences of the word "presented."

No Comments

No Known Activations