INDEX

Explanations

words related to physical injuries, specifically tearing

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "torn" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lain

-0.87

VERTISEMENT

-0.78

ITNESS

-0.77

perty

-0.75

racuse

-0.71

localhost

-0.71

ertodd

-0.69

arten

-0.66

ALT

-0.65

cius

-0.64

POSITIVE LOGITS

adoes

1.65

ados

1.12

adic

0.93

onite

0.78

 tires

0.77

 belts

0.76

neck

0.76

 apart

0.76

away

0.75

ado

0.74

Activations Density 0.021%

words related to physical injuries, specifically tearing

occurrences of the word "torn" in various contexts

No Comments

No Known Activations