INDEX

Explanations

words related to war-torn regions and humanitarian crises

oai_token-act-pair · gpt-3.5-turbo

references to war-torn regions and the consequences of conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

*/(

-0.73

HCR

-0.70

çīĪ

-0.68

uez

-0.66

herty

-0.65

 Burr

-0.65

veland

-0.65

kson

-0.64

 opin

-0.64

Vec

-0.61

POSITIVE LOGITS

ridden

1.12

seeking

1.10

prone

1.09

laden

1.09

proof

1.08

resistant

1.05

related

1.04

like

1.02

inducing

1.02

footed

1.01

Activations Density 0.113%

words related to war-torn regions and humanitarian crises

references to war-torn regions and the consequences of conflict

No Comments

No Known Activations