INDEX

Explanations

mentions of motives or reasons behind actions

oai_token-act-pair · gpt-3.5-turbo

references to motives behind actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

semble

-0.94

opy

-0.84

thumbnails

-0.84

alus

-0.81

ropolis

-0.78

ummer

-0.77

ogun

-0.72

hap

-0.72

redd

-0.71

mark

-0.70

POSITIVE LOGITS

 motives

1.15

 motive

1.10

 justifying

1.06

 motivations

1.03

 behind

1.00

 rationale

0.99

 motivation

0.94

why

0.87

 justify

0.87

 reasoning

0.82

Activations Density 0.070%

mentions of motives or reasons behind actions

references to motives behind actions or events

No Comments

No Known Activations