INDEX

Explanations

phrases related to political discussions, warnings, and decision-making

oai_token-act-pair · gpt-3.5-turbo

phrases associated with political or social commentary

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 flourishing

-0.59

 manif

-0.58

 Amon

-0.57

 mount

-0.56

 endeav

-0.55

 mounted

-0.55

 chained

-0.55

 junction

-0.55

lod

-0.55

 delegation

-0.55

POSITIVE LOGITS

ï¸ı

0.98

onto

0.82

ï¸

0.80

estic

0.79

their

0.77

efe

0.76

0.75

rough

0.74

them

0.74

orders

0.72

Activations Density 0.277%

phrases related to political discussions, warnings, and decision-making

phrases associated with political or social commentary

No Comments

No Known Activations