INDEX

Explanations

words related to processes and actions, such as 'process', 'implementation', 'evasion'

oai_token-act-pair · gpt-3.5-turbo

terms related to processes and outcomes 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 @ 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

etheless

-0.89

saw

-0.73

assetsadobe

-0.68

-0.65

Shares

-0.64

ccording

-0.63

dot

-0.60

asma

-0.60

ares

-0.59

 expects

-0.59

POSITIVE LOGITS

 obsolete

1.01

 moot

0.90

 accessible

0.82

solete

0.79

 worthwhile

0.79

 irrelevant

0.77

 easier

0.74

 redundant

0.73

 synonymous

0.72

 impossible

0.72

Activations Density 0.465%

words related to processes and actions, such as 'process', 'implementation', 'evasion'

terms related to processes and outcomes in various contexts

No Comments

No Known Activations