INDEX

Explanations

Keywords related to negativity or disapproval

oai_token-act-pair · gpt-3.5-turbo

words that describe negative or unpleasant qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Nanto

-0.73

tsky

-0.71

 Dynamics

-0.71

phrine

-0.71

ãĥ¼ãĥĨãĤ£

-0.70

 Guardians

-0.70

 briefs

-0.70

 Grayson

-0.69

SHIP

-0.69

*/(

-0.69

POSITIVE LOGITS

rep

1.09

avour

1.02

aved

1.02

heat

1.00

killed

1.00

olicited

0.97

oci

0.96

ut

0.93

ident

0.93

iders

0.92

Activations Density 0.011%

Keywords related to negativity or disapproval

words that describe negative or unpleasant qualities

No Comments

No Known Activations