INDEX

Explanations

recognizable names or terms

oai_token-act-pair · gpt-3.5-turbo

occurrences of names starting with 'H', 'T', 'C', 'R', or 'L'

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

NEC

-0.69

AMA

-0.61

-0.59

CTR

-0.59

 abdom

-0.59

rency

-0.58

 capital

-0.58

issance

-0.57

 Chilean

-0.57

ilty

-0.57

POSITIVE LOGITS

oshenko

0.79

atsu

0.76

iazep

0.76

emouth

0.75

otta

0.74

awk

0.74

ouk

0.73

assy

0.73

owski

0.72

oresc

0.71

Activations Density 0.136%

recognizable names or terms

occurrences of names starting with 'H', 'T', 'C', 'R', or 'L'

No Comments

No Known Activations