INDEX

Explanations

phrases related to public speaking or statements made in public settings

oai_token-act-pair · gpt-3.5-turbo

phrases related to reporting events or statements made in public contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 stalk

-0.57

 mould

-0.57

cept

-0.54

 gore

-0.54

 murd

-0.51

 untold

-0.51

imon

-0.51

 mamm

-0.51

 disproportion

-0.51

 corrupt

-0.51

POSITIVE LOGITS

 Thursday

1.14

 Wednesday

1.13

 Tuesday

1.12

 Friday

1.07

 yesterday

1.05

 Monday

1.02

Thursday

0.97

Tuesday

0.93

Wednesday

0.93

Monday

0.92

Activations Density 0.433%

phrases related to public speaking or statements made in public settings

phrases related to reporting events or statements made in public contexts

No Comments

No Known Activations