INDEX

Explanations

terms related to political figures pursuing information

oai_token-act-pair · gpt-3.5-turbo

verbs that indicate ongoing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

advertisement

-0.75

sequence

-0.72

Spoiler

-0.68

 glitch

-0.65

 destro

-0.65

Ö

-0.64

dash

-0.63

ammy

-0.63

wash

-0.62

pict

-0.62

POSITIVE LOGITS

luaj

0.75

................................................................

0.72

ingen

0.63

 diplomacy

0.63

 respecting

0.61

 Cuba

0.61

 Finland

0.59

ikk

0.59

 Ecuador

0.59

Emb

0.59

Activations Density 0.000%

terms related to political figures pursuing information

verbs that indicate ongoing actions

No Comments

No Known Activations