INDEX

Explanations

occurrences of the word "work" or related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to work, production, and project activities

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

cemic

-0.69

die

-0.68

 Cele

-0.65

rican

-0.65

iciency

-0.63

emic

-0.63

illy

-0.62

 stereotype

-0.60

 emph

-0.59

ilo

-0.59

POSITIVE LOGITS

Phase

1.06

 underway

0.99

 preparations

0.97

 Phase

0.89

 groundwork

0.88

Construction

0.87

 anew

0.86

 begun

0.83

 renovations

0.78

 phase

0.77

Activations Density 0.237%

occurrences of the word "work" or related terms

terms related to work, production, and project activities

No Comments

No Known Activations