INDEX

Explanations

statements related to percentages or numerical data

oai_token-act-pair · gpt-3.5-turbo

present tense and past tense forms of the verb "to be."

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

ĪĴ

-0.68

wake

-0.62

 congr

-0.62

ãĥķãĤ¡

-0.60

attach

-0.57

ossom

-0.55

 scars

-0.55

ful

-0.53

mone

-0.53

 marches

-0.52

POSITIVE LOGITS

 pegged

0.92

 estimated

0.84

 approximately

0.82

 around

0.80

 about

0.78

around

0.77

 hovering

0.74

 calculated

0.74

 believed

0.73

Activations Density 0.105%

statements related to percentages or numerical data

present tense and past tense forms of the verb "to be."

No Comments

No Known Activations

statements related to percentages or numerical data

present tense and past tense forms of the verb "to be."

No Comments

No Known Activations