INDEX

Explanations

IDs related to publishing, tracking, and research

oai_token-act-pair · gpt-3.5-turbo

mentions of specific identification labels or classification codes

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

theless

-0.76

 Silence

-0.72

iful

-0.68

bilt

-0.68

cffff

-0.67

uten

-0.66

sei

-0.65

 silence

-0.64

terday

-0.64

ussen

-0.64

POSITIVE LOGITS

aho

1.07

iots

1.06

LER

1.06

DEN

1.05

ENT

1.01

entity

1.00

irect

0.94

ACA

0.94

ictionary

0.93

FP

0.92

Activations Density 0.017%

IDs related to publishing, tracking, and research

mentions of specific identification labels or classification codes

No Comments

No Known Activations