INDEX

Explanations

words related to postponement or delays

oai_token-act-pair · gpt-3.5-turbo

terms associated with postponement and related concepts

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

 Duchess

-0.81

 Nanto

-0.77

Charge

-0.73

Reviewer

-0.73

DEN

-0.73

 Nerd

-0.73

REDACTED

-0.72

 Darling

-0.71

Accessory

-0.71

ãĥ¼ãĥĨ

-0.70

POSITIVE LOGITS

orns

1.07

rop

1.00

opes

0.99

Ľ

0.97

©

0.97

aped

0.95

agic

0.94

izoph

0.92

airs

0.91

»

0.90

Activations Density 0.061%

words related to postponement or delays

terms associated with postponement and related concepts

No Comments

No Known Activations