INDEX

Explanations

phrases indicating a consequence or implication

oai_token-act-pair · gpt-3.5-turbo

statements about implications or meanings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

thumbnails

-0.82

EStreamFrame

-0.79

oked

-0.67

uded

-0.62

Newsletter

-0.59

ics

-0.57

 keynote

-0.57

iliate

-0.56

ItemImage

-0.55

rete

-0.55

POSITIVE LOGITS

terday

0.94

hift

0.85

 goodbye

0.84

forth

0.67

antage

0.67

aucus

0.65

ãĥĨãĤ£

0.65

Ðº

0.64

bnb

0.62

ropy

0.61

Activations Density 0.027%

phrases indicating a consequence or implication

statements about implications or meanings

No Comments

No Known Activations