INDEX

Explanations

numerical values in a specific format, likely related to code or data processing

oai_token-act-pair · gpt-3.5-turbo

percentage signs and related numerical data

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ITED

-0.77

 compr

-0.74

 exerc

-0.72

 subpoen

-0.67

issued

-0.66

 Phant

-0.64

 fasc

-0.62

Rav

-0.61

 succession

-0.61

stru

-0.61

POSITIVE LOGITS

%%%%

0.97

ennial

0.83

reet

0.82

files

0.80

AppData

0.80

imate

0.80

username

0.80

urion

0.78

rowth

0.76

rax

0.75

Activations Density 0.034%

numerical values in a specific format, likely related to code or data processing

percentage signs and related numerical data

No Comments

No Known Activations