INDEX

Explanations

descriptions of individuals involved in incidents or events

oai_token-act-pair · gpt-3.5-turbo

references to individuals involved in incidents, particularly in legal or crime contexts

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

ipedia

-0.82

aeda

-0.76

 Bound

-0.67

ysical

-0.65

acebook

-0.63

olesterol

-0.63

 Profession

-0.63

 Poll

-0.61

 Publication

-0.61

ornia

-0.60

POSITIVE LOGITS

 suspects

0.77

's

0.76

hunt

0.75

stown

0.74

 testified

0.73

 belonged

0.73

 suspect

0.73

nels

0.71

liest

0.71

 victim

0.70

Activations Density 0.193%

descriptions of individuals involved in incidents or events

references to individuals involved in incidents, particularly in legal or crime contexts

No Comments

No Known Activations