INDEX

Explanations

phrases related to conditional statements or possibilities

oai_token-act-pair · gpt-3.5-turbo

present tense verbs that indicate conditions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pedia

-0.75

avascript

-0.74

Fish

-0.61

 Lives

-0.61

inas

-0.60

 Suite

-0.60

\",

-0.60

 ILCS

-0.59

interstitial

-0.57

\\\\\\\\

-0.56

POSITIVE LOGITS

 ever

1.13

 indeed

1.11

 successful

1.01

 truly

0.95

 unsuccessful

0.93

 unable

0.90

 anything

0.88

 unwilling

0.88

 lucky

0.86

 willing

0.85

Activations Density 0.122%

phrases related to conditional statements or possibilities

present tense verbs that indicate conditions or states

No Comments

No Known Activations