INDEX

Explanations

references to the television network "CW"

oai_token-act-pair · gpt-3.5-turbo

references to the CW (The CW television network) and its programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

iasis

-1.07

olulu

-0.84

heid

-0.79

ogy

-0.75

raged

-0.72

emort

-0.70

antly

-0.68

ority

-0.67

xtap

-0.67

uckland

-0.66

POSITIVE LOGITS

CW

0.93

ND

0.92

NF

0.88

NT

0.88

NC

0.86

trop

0.86

CW

0.85

RC

0.84

KI

0.83

HL

0.82

Activations Density 0.006%

references to the television network "CW"

references to the CW (The CW television network) and its programming

No Comments

No Known Activations