INDEX

Explanations

mentions of specific brands or companies

oai_token-act-pair · gpt-3.5-turbo

words related to products or meal replacements

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

STON

-0.77

side

-0.75

BILITIES

-0.74

CHR

-0.66

CTV

-0.65

iets

-0.65

STER

-0.65

jack

-0.64

head

-0.63

 Heist

-0.62

POSITIVE LOGITS

ropy

1.15

ral

1.11

ucky

1.07

ertain

0.98

rification

0.97

inel

0.96

imental

0.96

acles

0.94

acion

0.92

iment

0.90

Activations Density 0.048%

mentions of specific brands or companies

words related to products or meal replacements

No Comments

No Known Activations