INDEX

Explanations

names or terms related to people or places

oai_token-act-pair · gpt-3.5-turbo

repeated instances of specific names or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

TY

-0.86

starter

-0.79

ICLE

-0.74

tle

-0.73

space

-0.72

rd

-0.70

LIN

-0.70

RO

-0.70

lust

-0.69

COM

-0.69

POSITIVE LOGITS

velength

0.93

uthor

0.93

ÅĤ

0.82

Ã±

0.79

ruary

0.77

ãĤ´ãĥ³

0.77

Mae

0.76

quez

0.74

ÄŁ

0.74

zzle

0.73

Activations Density 0.042%

names or terms related to people or places

repeated instances of specific names or titles

No Comments

No Known Activations