INDEX

Explanations

words related to causality and outcomes

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate outcomes or consequences

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

tsky

-0.67

kov

-0.66

bones

-0.62

 spaced

-0.60

 Classical

-0.60

 craw

-0.59

borne

-0.58

 uncle

-0.58

Jeb

-0.57

ysis

-0.55

POSITIVE LOGITS

antly

0.85

rences

0.80

inally

0.79

ĸļ

0.78

uces

0.76

LEASE

0.75

INAL

0.75

uced

0.75

aneously

0.73

UE

0.73

Activations Density 0.038%

words related to causality and outcomes

phrases that indicate outcomes or consequences

No Comments

No Known Activations