INDEX

Explanations

phrases related to investigation and review processes

oai_token-act-pair · gpt-3.5-turbo

phrases indicating investigative or analytical actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

å§

-0.82

çĶ°

-0.76

ç·

-0.70

å¤©

-0.68

bucks

-0.66

uala

-0.66

åħī

-0.66

itol

-0.65

Sad

-0.64

EVA

-0.64

POSITIVE LOGITS

pse

0.82

 forensic

0.80

 eval

0.78

aminer

0.77

 psychiatrists

0.76

 interviewing

0.76

enhagen

0.76

 analys

0.73

 assessing

0.72

 dossier

0.71

Activations Density 0.388%

phrases related to investigation and review processes

phrases indicating investigative or analytical actions

No Comments

No Known Activations