INDEX

Explanations

phrases or sentences indicating problems or issues

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the word "problem."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

rib

-0.94

orks

-0.92

tein

-0.86

umber

-0.86

urses

-0.83

htaking

-0.81

vez

-0.80

collar

-0.80

rica

-0.78

cul

-0.78

POSITIVE LOGITS

 plag

0.99

Problem

0.99

 confronting

0.83

 solved

0.81

 solving

0.80

 facing

0.77

naires

0.75

 problems

0.74

 Problem

0.73

 problem

0.72

Activations Density 0.036%

phrases or sentences indicating problems or issues

repeated mentions of the word "problem."

No Comments

No Known Activations