INDEX

Explanations

words related to edibles or food items

oai_token-act-pair · gpt-3.5-turbo

words or phrases related to "ed" suffixes, indicating past actions or modified states

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

Story

-0.69

fruit

-0.64

Nun

-0.64

 sight

-0.61

EStreamFrame

-0.61

ãĥ¯ãĥ³

-0.60

Kou

-0.59

merce

-0.59

idine

-0.58

falls

-0.58

POSITIVE LOGITS

uce

1.21

monton

1.21

icts

1.17

uctive

1.14

nesday

1.13

ict

1.13

gew

1.11

uct

1.10

inburgh

1.05

aily

1.03

Activations Density 0.042%

words related to edibles or food items

words or phrases related to "ed" suffixes, indicating past actions or modified states

No Comments

No Known Activations