INDEX

Explanations

words related to allowing or giving permission

oai_token-act-pair · gpt-3.5-turbo

instances of the word "let" in varying forms and 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

 Languages

-0.79

availability

-0.68

lihood

-0.68

 millenn

-0.67

oppable

-0.62

liest

-0.62

DAQ

-0.60

Interstitial

-0.60

bage

-0.59

Zen

-0.58

POSITIVE LOGITS

tered

1.34

tering

0.98

 slip

0.90

 loose

0.89

icia

0.86

go

0.81

us

0.77

itia

0.75

him

0.71

onomy

0.66

Activations Density 0.044%

words related to allowing or giving permission

instances of the word "let" in varying forms and contexts

No Comments

No Known Activations