INDEX

Explanations

instances of potential conflicts and conflicts of interest

oai_token-act-pair · gpt-3.5-turbo

words associated with various types of conflicts

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

trak

-0.81

GC

-0.76

************

-0.75

GV

-0.73

Stud

-0.72

girls

-0.71

slow

-0.70

AA

-0.69

 Pione

-0.68

DT

-0.68

POSITIVE LOGITS

 conflicts

0.93

 conflict

0.82

 unresolved

0.81

 between

0.78

 situations

0.72

 arising

0.71

 resolves

0.71

 conflicted

0.69

 Conflict

0.69

 misunderstand

0.69

Activations Density 0.017%

instances of potential conflicts and conflicts of interest

words associated with various types of conflicts

No Comments

No Known Activations