INDEX

Explanations

words related to titles or positions within a hierarchy

oai_token-act-pair · gpt-3.5-turbo

phrases indicating positions, titles, or roles someone holds

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

IFrame

Link

Negative Logits

raints

-0.75

venants

-0.73

 reperto

-0.73

ieties

-0.73

imon

-0.72

ities

-0.72

ento

-0.72

igun

-0.71

isons

-0.70

 formulations

-0.69

POSITIVE LOGITS

 medi

0.77

 owning

0.69

 defending

0.68

 facilitating

0.67

 educating

0.67

 course

0.66

 assisting

0.66

 pretending

0.65

 confirming

0.65

 usher

0.65

Activations Density 0.167%

words related to titles or positions within a hierarchy

phrases indicating positions, titles, or roles someone holds

No Known Activations