INDEX

Explanations

instances where the text expresses strong opinions or evaluations about something

oai_token-act-pair · gpt-3.5-turbo

conditional statements or phrases that suggest scenario-based reasoning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ãĤª

-0.75

roid

-0.64

rosse

-0.62

%);

-0.62

FTWARE

-0.61

akia

-0.60

 GOODMAN

-0.59

heimer

-0.59

holm

-0.58

yss

-0.58

POSITIVE LOGITS

fy

0.87

tar

0.77

edi

0.73

 unchecked

0.69

you

0.67

rame

0.67

 anything

0.66

acebook

0.65

iably

0.64

FIN

0.62

Activations Density 0.075%

instances where the text expresses strong opinions or evaluations about something

conditional statements or phrases that suggest scenario-based reasoning

No Comments

No Known Activations