INDEX

Explanations

time-related phrases or events

oai_token-act-pair · gpt-3.5-turbo

references to specific historical or chronological periods

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

avorite

-0.67

 Hits

-0.65

 suites

-0.64

 Graveyard

-0.64

cyt

-0.63

 bluff

-0.63

 subst

-0.63

tad

-0.63

 dent

-0.60

 bets

-0.60

POSITIVE LOGITS

SPONSORED

0.93

cture

0.82

å£«

0.81

pan

0.81

zone

0.79

å¥³

0.76

cia

0.76

cale

0.73

cens

0.73

frames

0.73

Activations Density 0.051%

time-related phrases or events

references to specific historical or chronological periods

No Comments

No Known Activations